Enhancing Query Expansion Method Using Word Embedding

Abstract

Recently, many query expansion methods have been proposed to improve the results of search applications. However, many of these search applications still lack better results and many attributed due to query expansion issues. This paper enhanced the query expansion method based on unigram model with Okapi BM25and word embedding using Glove. A Glove model captured the semantic similarity by mapping various words based on unigram with Okapi BM25 results. The results indicate that our proposed method based on Glove model word embedding can significantly improve query expansion methods using Arberry dataset.